Aviz Networks Blogs

Just finished an online seminar on how AI-powered deep network observability can help government agencies meet high standards of security and regulation. It was really striking how similar the discussion was to what actual frontline teams delivering essential services are experiencing.  Some Observations:  Governments are evolving and growing quickly in the digital space. They have to bring all of that into sync with standards and regulations set out by groups like NIST, CMMC, and FISMA. Old-school monitoring tools can insert holes and cause a whole lot of extra work  Seeing the whole network and how it all interacts is the first step to implementing Zero Trust and catching problems early AI can do the boring bits of compliance and produce audit-ready compliance reports Cross-layer telemetry correlation helps with the rapid detection and elimination of problems Keeping your data on-premises will remain the best way to keep it secure and meet your compliance obligations. I...

Just a couple of days ago I heard this podcast titled Deciding the Future Building Networks for the AI Era and I am just really surprised and rethinking how important networking is now. What I liked the best is that it doesn’t stay conceptual. It really explains how AI is transforming the core of infrastructure particularly in networks. Just a couple of bits that I found interesting: Networking has evolved from being mere support infrastructure to being the core of scaling AI systems The shift is no longer slow it’s quick and exponential Transitioning from a single network to specialized networks (training, inference, etc.) Decisions now may be very difficult to undo in the future Open and adaptable architectures are now a requirement, not a choice Network traffic behaviors are starting to change even if AI isn t directly being used My takeaways from that Something that really stuck with me: AI is no longer something you can tack on.It has to be integrated into the core, particularly i...

Yeah, I got really into DPI metadata collection as a type of application observability last week and how it can technically do this without having to decrypt the traffic. What was interesting was how workable it was for modern encrypted networks. Practical points: Over 95% of network data is now encrypted making conventional application monitoring techniques ineffective. Packet capturing records the network traffic but does not reveal the functioning of an application Parsing of metadata provides formatted data for HTTP, DNS, TLS, QUIC, DHCP, and RTP. Faster issue isolation by separation at DNS, TLS, HTTP or transport layer. Metadata can be searched to meet compliance standards and the payload data remains hidden. Just to clarify deep visibility does not require payload decryption. Structured metadata, on the other hand, provides adequate data about how an application works without compromising the vital privacy bounds. What I really got from it is that observability is so much better ...

  Issue Reported by Customer Device: AS9716-32D Problem: Device became unreachable — SSH and console access failed. Customer Action Taken Power cycle performed Device recovered and became reachable again Customer Concern Why did the device suddenly become unreachable? Why was a critical temperature alert triggered at 66°C while the platform shows higher thresholds? The logs told the story During log analysis, we identified thermal warnings followed by a critical shutdown event . Temperature Spike Logs Feb 3 18:55:27 WARNING pmon#thermalctld: Temperature of CPU Package Temp changed too fast, from 32.0 to 66.0 Feb 3 18:55:27 WARNING pmon#thermalctld: Temperature of CPU Core 0 Temp changed too fast, from 32.0 to 66.0 Feb 3 18:55:46 CRIT - Monitor CPU Temp, temperature is 66.0. Temperature is over 66.0. Need shutdown DUT. Immediately after this event:  Device stopped responding SSH and console access unavailable After the reboot: all clear After the power cycle, system tempera...

  What If the Real AI Bottleneck Isn’t Compute but the Network? I recently went through a detailed discussion exploring how networks are evolving in the AI era, and what stood out was how fundamentally the conversation challenges traditional thinking about infrastructure. Some practical observations: • AI is driving exponential changes in compute, applications, and network requirements • Traditional step-by-step network scaling models no longer hold true • Network design must now align closely with rapidly evolving chip and system architectures • Distinguishing between networks for AI and AI for networks is critical • Traffic patterns are changing significantly, even in environments without GPUs One misconception addressed during the discussion was that networks are simply plumbing. In reality, network decisions today directly determine whether AI strategies succeed or fail over the next decade. My biggest takeaway is that networks are no longer a supporting layer. They are a strat...

  I recently went through a detailed announcement describing how AI infrastructure is evolving into a fully managed,  multi-tenant platform. What stood out was how grounded the discussion was in real operational challenges across  GPU environments. Some practical observations:  • Building GPU clusters is only the first step; consistent operations remain a major challenge  • AI infrastructure requires tight alignment between compute, networking, storage, and orchestration layers  • Multi-tenant environments need strong isolation and policy-driven controls  • Lifecycle automation from deployment to ongoing operations is critical for scale  • Unified observability across infrastructure and workloads improves troubleshooting and efficiency One misconception addressed in the discussion was that deploying AI infrastructure is primarily about hardware.  In reality, long-term success depends on how well the entire stack is integrated and operated. My...